Does Automated Collateral Evaluation Lower Mortgage Credit Risk Relative to Home Appraisal Valuations?

Kadiri Karamon & Douglas A. McManus

The Journal of Structured Finance2022https://doi.org/10.3905/jsf.2022.1.138article
AJG 1ABDC C
Weight
0.40

What the paper says

Automated valuation models (AVMs) are increasingly being used as a substitute for home appraisals in mortgage origination. This article examines whether there are differences in the credit risk of mortgages originated using AVMs relative to traditional appraisals. This question is explored through modeling the conditional default rates of loans originated through Freddie Mac’s automated collateral evaluation (ACE) program relative to those originated with appraisals. We show that ACE loans have about a 9.6% lower default risk in comparison to otherwise similar loans originated with appraisals. This superior default performance can be explained by the ACE decision logic employing data beyond that found in a traditional appraisal and the more efficient utilization of available information through statistical modeling. These findings are limited to Freddie Mac’s use of its AVM, the Home Value Explorer®, and the algorithmic decision logic in the ACE program.

2 citations

Open paper page →

Cite this paper

https://doi.org/https://doi.org/10.3905/jsf.2022.1.138

Or copy a formatted citation

@article{kadiri2022,
  title        = {{Does Automated Collateral Evaluation Lower Mortgage Credit Risk Relative to Home Appraisal Valuations?}},
  author       = {Kadiri Karamon & Douglas A. McManus},
  journal      = {The Journal of Structured Finance},
  year         = {2022},
  doi          = {https://doi.org/https://doi.org/10.3905/jsf.2022.1.138},
}

Paste directly into BibTeX, Zotero, or your reference manager.

Flag this paper

Does Automated Collateral Evaluation Lower Mortgage Credit Risk Relative to Home Appraisal Valuations?

Flags are reviewed by the Arbiter methodology team within 5 business days.


Evidence weight

0.40

Balanced mode · F 0.40 / M 0.15 / V 0.05 / R 0.40

F · citation impact0.14 × 0.4 = 0.06
M · momentum0.80 × 0.15 = 0.12
V · venue signal0.50 × 0.05 = 0.03
R · text relevance †0.50 × 0.4 = 0.20

† Text relevance is estimated at 0.50 on the detail page — for your query’s actual relevance score, open this paper from a search result.